Seismic Vulnerability Analysis For Bridge Network And Algorithm Improvement

Bridges, in particular, play a vital role in modern transportation:they are the key nodes of various road networks. Under the influence of external environment factors (e.g. earthquakes), bridge damage may lead to partial or even the whole network interruption causing a negligible impact on the reliability of the traffic network. However, the results of seismic micro-zoning by CPSHA are not usually same with that of seismic regionalization map, but most common bridges’strength are designed according to the intensity paremeters from the seismic regionalization map. As a result, the discrepancy between CPSHA and seismic regionalization map would result that the strength reserves of bridegs in same seismic zone vary with different seismic micro-zone. Therefore, to evaluate the road network reliability, the regional seismic hazard and the seismic bridge vulnerability are analyzed in this thesis.CPSHA (China Probabilistic Seismic Hazard Analysis) method is chosen as the most appropriate to evaluate the potential seismic hazard of the sites in the network located in the southwest China. The obtained calculation results are further used to analyze the seismic vulnerability of individual bridges and then generate the fragility curves for all the bridges. Thus, given an earthquake scenario, seismic vulnerability for all damage states can be calculated. After this, the results of the component level analysis are extended to the network level and the problem of connectivity between two locations is analyzed. On the basis, the Target_Order algorithm is proposed. A metro network is selected as an example to illustrate the considerable advantages of the proposed algorithm against traditional algorithm in terms of computational accuracy and efficiency. The improved solution algorithm, as proved, is more suitable to be applied to connectivity reliability evaluations of large-scale complex networks. Moreover, a calculation model of the’path cost’is defined to identify the best path between any two given places after an earthquake using the shortest path algorithm.The results show that the damage probabilities of bridges located in No.10,37,39,40 site are larger than the others’with exceeding probability of 0.1 in 50 years. In addition, it is found that the upper and lower bounds of the network connectivity reliability will approach each other and tend to be stable when the number of network states increases and the speed of approximation is affected by the level of earthquake risk. These results can provide design parameters for the engineering projects in southwest China and can assist bridge managers and government officials in making bridge optimum maintenance strategies and rapid decisions with regard to distributing the available human and material resources to the disaster centre in neral-real time, under post-earthquake conditions.